Improving Data-based Wind Turbine Using Measured Data Foggy Method

The purpose of this paper is to improve the modeling of the data-driven wind turbine system that receives data from noise signals. Most of the data on industrial systems is noisely and data noise is inevitable and natural. The method and idea proposed in this paper, Data Fogging, significantly reduce the impact of noise on data-driven wind turbine system modeling, which is the basis of this method of changing the acceptable range of measured data. It will not eliminate noise in the system data, but will significantly reduce their impact on detection, modeling and fault detection. In this paper, the proposed method of artificial neural networks is used as a case study for modeling of wind turbine power transmission system of Kahak wind power plant. Noise and noisely information data applied to the system, modeling performed, and the results are reviewed. The results of the simulation presented in the tables and figures below illustrate the very good and accurate performance of the data fogging method in eliminating the impact of noise on system modeling, as the case of the wind turbine system proves. In fact, the impact of this method on real systems analysis is noise, because the noise in industrial systems has an impact on system analysis and greatly affects the accuracy of system identification, modeling and fault detection. Simulations of the method presented in this paper have been performed in MATLAB software. The reason for choosing this software to perform the simulation is that it is very powerful and reliable. In the research section of this paper, first a method and idea is presented and control analysis is performed and then in the software simulation section of this paper the proposed method is implemented and simulated on a real study system (wind turbine of Kahak wind power plant).